Sensor Systems Laboratory

Inventing new technology for sensing, robotics, wireless power, and medical devices

Joshua R. Smith, Ph. D.
Milton and Delia Zeutschel Professor
Allen School of Computer Science and Engineering, University of Washington
Department of Electrical and Computer Engineering, University of Washington

Brain Computer Interface for Robots

BCI Researchers

Rajesh N. Rao, Principal Investigator

Joshua R. Smith, co-Principal Investigator

Lillian Chang, Sensor Systems Postdoc

Mike Chung, Neural Systems Grad Student

Griffin Nicoll, Sensor Systems Undergrad

Vibinash Thomas, Sensor Systems Undergrad

Summary

We are been collaborating with the Neural Systems Lab to create a Brain Computer interface (BCI) for Hobbes (our PR2 robot), which will allow the robot to perform remote tasks. The system consists of our PR2 robot, an electrode cap for sensing brainwave frequencies, and graphical user interface for controlling the robot remotely. This system will allow a severely paralyzed patient to use a robot as a proxy to perform a task by issuings commands to the robot without the need of physical movement.

Current Projects

Grasping Objects Interaction

Shell Game Interaction

More Information

Because the user is located in a remote room wearing the cap, a graphical interface is necessary for the user to interact with the robot. The user monitors the robot's state through a live video feed on the screen. This screen also displays potential commands for the robot to perform.
BCI is based on the Steady State Visually-Evoked Potentials (SSVEP) paradigm. This type of BCI operates by exposing the user to oscillating visual stimuli. Electrical activity corresponding to the frequency of this oscillation (and its multiples) can be measured from the occiptial lobe of the brain located at the back of the skull. The user issues a command by choosing a stimulus (and therefore a frequency) to pay attention to. The BCI measures the corresponding EEG activity and attempts to infer from it the command the user has chosen for execution.

Pre-Touch Publications

Pretouch Sensing for Manipulation, Liang-Ting Jiang, Joshua R. Smith, Robotics: Science and Systems, Workshop: Alternative Sensing Techniques for Robotic Perception, Jul 11-12, 2012.

Seashell Effect Pretouch Sensing for Robotic Grasping, Liang-Ting Jiang, Joshua R. Smith, IEEE International Conference on Robotics ans Automation (ICRA), May 5-12, 2012.


Seashell Effect Pretouch for Robot Grasping, Liang-Ting Jiang, Joshua R. Smith, IROS 2011 PR2 Workshop.